ASM Assembly Automation Ltd v Aurigin Technology Pte Ltd and Others

Tan Lee Meng J:

[1] The plaintiff, ASM Assembly Automation Ltd (“ASM”), a Hong Kong company that is in the business of manufacturing and selling semiconductor assembly and packaging equipment, is the registered proprietor of Patent No 104354 (the “patent”), which was granted on 30 December 2005. The priority date of the patent, which relates to an apparatus and method for automatically placing an array of solder balls onto a substrate, such as a ball-grid array (“BGA”) substrate, is 8 January 2002. The said patent was granted pursuant to s 29(1)(c) and s 29(5) of the Patents Act (Cap 221, 2002, Rev Ed) (“the Act”), relying on its corresponding US patent application (US Patent No 6,766,938B2).

[2] ASM alleged that the 1 defendant, Aurigin Technology Pte Ltd (“Aurigin”), a Singapore company, infringed the patent and that the 2 defendant, Mr Lim Ee Teoh (“Mr Lim”), and the 3 defendant, Mr Tam Wing Wah (“Mr Tam”), who are Aurigin’s directors, directed, authorised, counselled or procured Aurigin to infringe the patent. Aurigin’s allegedly infringing product, the AU800, is its flagship product. Invented by Mr Lim and Mr Tam, together with one Mr Ng Boon Chew, the AU800 is an automated BGA solder ball placement machine which houses a key solder ball placement module based on its patented solder ball placement module design. The said module was patented in Singapore (Patent No 111523) on 30 April 2007, with a priority date of 14 October 2002, and in the United States (Patent No 10/531,611) on 2 December 2008. The first 2 sets of the AU800 were sold by Aurigin on 19 November 2003.

[3] ASM sought a declaration that its Singapore patent is valid and that the defendants’ AU800 infringed claims 1-4, 6-7 and 11-12 of its amended patent. ASM also sought an injunction to restrain the defendants from infringing its Singapore patent and an inquiry as to damages or alternatively an account of profits made by Aurigin and its directors as a result of the alleged infringement of its patent.

[4] Aurigin denied having infringed ASM’s patent and sought an order that ASM’s patent be revoked for lack of novelty and inventiveness. The defendants pleaded that even if ASM’s patent is valid, they are entitled to rely on s 71(1) of the Act, which concerns the defence of having made serious and effective preparations to manufacture and to offer to sell the AU800 before the priority date of ASM’s patent. They also pleaded the defence of innocent infringement and asserted that ASM is not entitled to any damages or account of profits prior to 6 December 2006.

[5] Aurigin also sought an injunction to restrain ASM from threatening their customers with legal proceedings for infringing ASM’s patent. Finally, Aurigin counterclaimed against ASM for having made groundless threats of infringement and sought an inquiry as to damages suffered.

[6] ASM’s expert witness was Assoc Prof Khong Poh Wah (“A/P Khong”) from the Nanyang Technological University (“NTU”) School of Mechanical and Aerospace Engineering. ASM also called Mr Robert Faber (“Mr Faber”) , a patent lawyer and a partner of Ostrolenk, Faber, Gerb & Soffen LLP, the American intellectual property firm that was responsible for prosecuting ASM’s United States Patent, to testify as an expert witness but he is, by his own admission, not a person skilled in the art. In Ng Kok Cheng v Chua Say Tiong [2001] 3 SLR 487, a case concerning the making of a lock, it was held that a patent attorney, who was called by the defendant as an expert, was not an expert as he was not a person skilled in the art of making locks. Likewise, Mr Faber’s evidence as a patent attorney is not relevant to issues pertaining to the views of a person skilled in the art.

[7] Aurigin called two expert witnesses to testify. The first was Assoc Prof Tay Meng Leong (“A/P Tay”), who, like A/P Khong, is from the Nanyang Technological University School of Mechanical and Aerospace Engineering. Aurigin’s second expert, Mr Dennis Miller (“Mr Miller”), was, until his retirement in 2006, the senior director and vice-president of Motorola Inc in Illinois. Mr Miller has rather impressive credentials as an inventor and innovator. A mechanical engineer, he holds 11 patents, 5 of which relate to BGA technology. His evidence was most useful to the court.

[8] As the patent relates to BGA technology, the following primer on this subject by Aurigin’s expert witness, Mr Miller, in his affidavit of evidence-in-chief (“AEIC”) at [18] is helpful:

[9] As the size of IC chips became smaller, newer technology was required to handle smaller solder balls. For instance, flux contamination, which was not a problem with larger solder balls, posed a serious problem for smaller solder balls. To negate this problem, Motorola developed equipment with a flux plate that contained flux. The solder balls on the pickup tool were partially immersed into the flux on the flux plate before being positioned on the substrate. Motorola obtained a patent for this invention on 8 February 1994.

[10] In due course, solder balls became so miniscule that a table spoon might contain a million solder balls. The significant reduction in the size of solder balls created new problems. For instance, the pickup tool had difficulties picking up solder balls from the reservoir of balls. Debris, flux contamination and machining variations aggravated the situation. Solder balls were also subject to oxidation and damage. A solution to the problems posed by the miniscule size of solder balls was to separate the major elements (solder ball reservoir, solder ball filling, picking up of solder balls, substrate fluxing) into independent processes. Thus, the pickup head would be presented with balls in a template, with one ball per vacuum hold and the filling of the template would be separated from the pickup head. The new fill process would involve flowing many solder balls over an array of apertures by running a ball reservoir over the apertures on the template.

[11] Many solutions to the difficulties posed by the downsizing of solder balls were patented. As for ASM’s patent, the abstract was in the following terms:

[12] The claims under ASM’s patent, as amended during the trial, are as follows:

[13] As ASM sought a declaration that its Singapore patent is valid while Aurigin sought a declaration that it be revoked, the validity of ASM’s patent will first be considered.

[14] For the purpose of determining the validity of ASM’s patent, reference must be made to s 13(1) of the Act, which provides that a patentable invention is one that satisfies the following conditions:

As condition (c) is not in issue, what matters is whether the invention is new and whether it involves an inventive step.

[15] Section 14 of the Act provides that an invention “shall be taken as new if it does not form part of the state of the art”. Section 14(2), which explains the meaning of “state of the art”, provides as follows:

[16] In regard to the issue of novelty, in Trek Technology (Singapore) Pte Ltd v FE Global Electronics Pte Ltd (No. 2) [2005] 3 SLR 389, Lai Kew Chai J explained as follows:

[17] As issues of novelty and inventiveness are viewed in relation to the prior art, the following summary by Mr Miller of the prior art at the time ASM obtained its patent is helpful:

[18] Also included but not explained in Mr Miller’s list of prior art are Cobbley and Fjelstad. In regard to these two, A/P Tay explained as follows:

[19] The defendants cited works by Oguro, Matsushita (Kasai), Tondelli and Stumpe to support their assertion that ASM’s claim 1 is not new. ASM retorted that all these works, with the exception of Oguro, had been considered during the examination of the corresponding US patent application and they had been found to be irrelevant to the patentability of the claims in its US patent, which mirrors the claims in the Singapore patent. As for Oguro, ASM contended that it is not any more relevant than the works already considered by the US examiner before the US patent was approved.

[20] ASM relied on A/P Khong’s conclusion in his opinion in his AEIC, where he states at [58] as follows:

[21] What matters is not an expert’s conclusion but the reasons for the conclusion. In this respect, A/P Khong’s evidence at the trial showed that he was not familiar enough with the prior art. On far too many occasions, he had to retract his views during the trial. A/P Khong explained what ASM’s invention teaches in the following terms:

[22] A/P Khong did not explain cogently what was novel about ASM’s patent. To begin with, his assertion that ASM’s patent is different from what had been disclosed in the prior art because it had a cover or lid on the box could not be taken seriously. When cross-examined, he said as follows:

[23] When cross-examined further, A/P Khong was forced to admit that he was absolutely wrong. The relevant part of the cross-examination is as follows:

[24] A/P Khong next pointed out that ASM’s patent is new because it taught a tilting mechanism. Here again, he was outflanked during cross-examination and he had to concede that this had been previously taught by others. The relevant part of the cross-examination is as follows:

[25] A/P Khong also asserted that ASM’s patent shows that solder balls flow downwards at a faster rate due to gravitational pull. What followed during cross-examination shows how unoriginal this idea of gravitational pull was:

[26] Subsequently, A/P Khong conceded that the use of gravitational force had already been made known in the prior art. The relevant part of the proceedings is as follows:

[27] A/P Khong next tried to distinguish ASM’s patent from the prior art on the ground that it teaches tilting in both directions. However, when cross-examined on the prior art, A/P Khong again had to concede once again that he was wrong. The relevant part of the proceedings is as follows:

[28] I now turn to Mr Miller’s evidence, which showed his mastery of the subject, with the reasons for his conclusions explained very clearly and authoritatively. Mr Miller asserted that claim 1 of ASM’s patent is not new because prior art documents, namely Tondelli, Kasai, Stumpe and Oguro individually disclosed all the elements claimed in claim 1. He explained in his AEIC at [34] as follows:

[29] As for the remaining words in ASM’s amended claim 1, namely an “apparatus ….thereby tilt it”, Mr Miller opined that it lacks novelty in view of Kasai and Oguro.

[30] In relation to ASM’s claim 2, Mr Miller stated that it is not novel in view of Oguro, which describes a tilting mechanism which is operative to tilt the mask and solder balls multiple times.

[31] As for ASM’s claim 3, Mr Miller said that it is not novel in view of Oguro and Kasai. While Oguro discloses rotational angles and the controllability of the moving speed of the container, Kasai describes a tilting mechanism which can control the rotational angle of the container by “adjusting pivoting directions and angles of the tilting mechanism” and it is clear that the moving speed of the container is controlled by a speed controller, which is described.

[32] Turning to ASM’s claim 6, Mr Miller rubbished any claim to novelty in view of Tondelli and Kasai. He said that Tondelli discloses that the holes of drilled areas are the same as the apertures in the template and that the diameter of these holes is slightly greater than the diameter of the solder balls. Furthermore, it is evident from Kasai’s illustration that the diameter of the apertures is larger than the size of a solder ball.

[33] As for ASM’s claim 7, which refers to an “additional pick-and-place device” to pick the solder balls in the apertures, while retaining their respective positions to be placed onto corresponding positions of solder pads on a substrate, Mr Miller stated that it lacks novelty in view of Tondelli and Kasai, who both disclose this device.

[34] Mr Miller also had no doubt that ASM’s claim 11, which highlights the fact that the “container is closed at the top to reduce oxidation of the solder balls and open at the bottom for direct entry of the balls to the positioning member substantially throughout the bottom of the container” lacks novelty. He explained that Stumpe defines the ball bin as being solid and having a closed top and open bottom.

[35] In regard to claim 12, which describes the location of the apertures as being near one end of the positioning member, Mr Miller asserted that this is also not novel. After all, Tondelli illustrates the location of the apertures according to this definition.

[36] I unreservedly accept Mr Miller’s explanations as to why ASM’s patent lacked novelty.

[37] As for the question of inventiveness, s 15 of the Act provides that an invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art, having regard to any matter which forms part of the state of the art by virtue only of section 14(2). In Windsurfing International Inc v Tabur Marine [1985] RPC 59 at 73, Oliver J offered the following oft-cited elucidation of inventiveness:

[38] Although the Windsurfing test has been subject to much scrutiny, it was approved by the Court of Appeal in First Currency Choice Pte Ltd v Main-Line Corporate Holdings Ltd [2008] 1 SLR 335 (“First Currency”).

[39] It is noteworthy that in Unilever v Chefaro [1994] RPC 567, Jacob J stated at p 580 as follows:

[40] As for expert evidence on the question of inventiveness, in SmithKline Beecham plc v Apotex Europe [2005] FSR 23, Jacob LJ said at pp 52-53 as follows:

[41] When dealing with obviousness, one is entitled to make a “mosaic” out of relevant documents if it can be put together by an unimaginative man with no inventive capability (see Technograph v Mills & Rockely [1972] RPC 346). In the present case, ASM’s invention “relates to an apparatus and method for automatically placing an array of solder balls onto a substrate, such a ball-grid array (BGA) substrate” and ASM’s patent describes some shortcomings in the prior art, including the following:

[42] It is thus not surprising that the summary of ASM’s invention states that “it is an objective of the present invention to provide an improved apparatus for automated placement of solder balls onto a substrate so that damages and/or oxidation of solder balls are reduced as compared to the prior art”.

[43] ASM’s expert witness, A/P Khong thought that there were three aspects of inventiveness of ASM’s patent. When cross-examined, he stated as follows:

[44] In regard to inventiveness, A/P Khong tied himself up in knots. As for the first inventive feature claimed by A/P Khong, namely multiple opportunities to fill solder balls, Aurigin’s counsel, Mr Low rightly pointed out that this is not a real advantage when others, including the holders of patents constituting the prior art, teach the finishing of the job with only one stroke. A/P Khong agreed as he stated as follows during cross-examination:

[45] Apart from A/P Khong, ASM’s own inventor, Mr Cheng Chi Wai, said during cross-examination that the most efficient method to fill solder balls is to utilize the tile mechanism once. His exact words are “For the best efficiency, rotate one time is the best, just one time tilting”.

[46] No one from ASM could articulate the inventiveness of a double tilting mechanism other than to say that it increases the opportunity for filling solder balls, a feature which is irrelevant if a machine is better because it does not require a second or third tilt. When Aurigin’s counsel, Mr Low Chai Chong asked A/P Khong about Aurigin’s claim that its AU800 does its job of filling solder balls in just one stroke, the latter conceded that if it does, it would be a superior machine. The relevant part of the cross-examination is as follows:

[47] A/P Khong said that he had not seen the AU800 and had not carried out any tests on this machine. The efficiency of Aurigin’s AU800 was tested and confirmed by A/P Tay, who stated as follows in his AEIC at [32]:

[48] In short, ASM’s patent may be viewed as a disadvantageous modification of the closest prior art in that it requires more than one opportunity to fill solder balls. In this regard, it ought to be borne in mind that in EPO Case Number T119/82, the EPO held that an invention which is the result of a foreseeable disadvantageous modification of the close prior art, which the skilled person can clearly predict and correctly assess, and this predictable disadvantage is not accompanied by an unexpected technical advantage, the claimed invention would not involve an inventive step.

[49] Turning to the second of the alleged inventive steps, namely minimization of damage to solder balls, Aurigin’s expert witness, Mr Miller could not see how this involved an inventive step. During re-examination, he opined as follows:

[50] As for the third of the alleged inventive steps, namely reduction of oxidation, A/P Khong was not very coherent. The relevant part of the cross-examination is as follows:

[51] What was troubling was the following exchange between Aurigin’s counsel and A/P Khong, during which it was clear that there was very little difference between the rate of oxidation in the systems disclosed in the prior art and that disclosed in ASM’s patent. A/P Khong repeated his discredited assertion that the cover of the container in ASM’s patent made the difference. When cross-examined, he stated as follows:

[52] What was also rather startling was A/P Khong’s response to Mr Low’s carefully crafted question on the practical value of the alleged 3 inventive steps in ASM’s patent. The relevant part of the proceedings is as follows:

[53] The defendants’ expert witness, A/P Tay, who was of the view that ASM’s patent showed no inventiveness, concluded in his AEIC (at [50] - [52]) as follows:

[54] A more detailed explanation as to why ASM’s patent disclosed no inventive steps was furnished by Mr Miller. He was, unlike A/P Khong, clearly non-partisan for when cross-examined, he made it absolutely clear that there was nothing inventive about both ASM’s patent and Aurigin’s patent. The relevant part of the cross-examination is as follows:

[55] As the invalidity and revocation of Aurigin’s patent were not pleaded and are not issues before the court, only the validity of ASM’s patent needs to be considered in this judgment. In regard to ASM’s patent, Mr Miller backed his assertion that it was not inventive with forceful submissions. With respect to claim 1 of the said patent, he explained that it is obvious because of Ooroku. Although ASM’s counsel submitted that Mr Miller did not appreciate that a very small step or a very slight advance could be an inventive step, the fact remains that Mr Miller minced no words when he declared that the ASM invention “is nothing but a workshop variation of Ooroku because the trailing edge of the container during the removal stroke in the ASM invention is still designed to move over the positioning member, thus causing damage to partially engaged solder balls”. Innumerable cases illustrate that such workshop improvements are insufficient to establish inventiveness. For instance, in Shaw v Burnet & Co [1924] 41 RPC 432, it was held that the addition of a piece of plastic to an ordinary bag veil to prevent it from slipping off a hat was a mere workshop improvement that did not involve any invention. Another instructive case is Curtis & Son v Heward & Co (1923) 40 RPC 53, where it was held that an alleged invention that made improvements to abdominal supports for medical purposes was not patentable as it was an ordinary workshop improvement of a well-known type of apparatus. I had no doubt whatsoever that with his wealth of experience in BGA technology, Mr Miller is in a position to explain what is a workshop variation of existing prior art in BGA technology.

[56] In regard to claim 2, which includes an additional operability of the apparatus to rotate the container and template multiple times when the container is directly over the apertures of the template to fill the apertures with solder balls, Mr Miller said “it would have been obvious to combine the operability disclosed in Inoue with any apparatus to achieve the objective of repeatedly spreading over the positioning member”. He added as follows:

[57] As for claim 3 of ASM’s patent, which concerns the further inclusion of “an additional controllability of the rotational angles of the container and/or the moving speed of the container to optimize the efficiency of the apparatus”, Mr Miller had no doubt that it is obvious to combine the operability disclosed in Oguru and/or Kasai with any apparatus to achieve the objective of repeatedly spreading over the positioning member.

[58] With respect to claim 4 of ASM’s patent, which claims an additional element of a positioning rail and motor whereby the container is positioned and driven, Mr Miller said that it is obvious in view of Lim Ee Teoh, which teaches the element of a positioning rail in the form of tracks. Furthermore, the additional element being the external mechanical means for driving the container is standard engineering practice as a mechanical means is always required for any motion.

[59] Turning to an analysis of claim 6 of ASM’s patent, which defines the apertures in the template as each being larger than the size of a solder ball, Mr Miller opined that Inoue defines the diameter of the apertures as being larger than the diameter of the solder ball and it would have been obvious to define the diameter of the apertures as being larger than the diameter of the solder balls since the objective of the apparatus is to position the solder balls into the apertures.

[60] Claim 7 of ASM’s patent, which refers to an “apparatus according to claim 6, wherein solder balls captured in the plurality of apertures are removable by a pick-and-place device while retaining their respective positions and placed onto corresponding positions of solder pads on the substrate”, was also regarded as an obvious claim by Mr Miller. He said that it would have been obvious to combine the device described in Tondelli or Kasai with any other document because removing the solder balls while retaining their respective positions to be placed on a target substrate is the next step of the process and this pick-and-place device had already been disclosed in Tondelli and Kasai.

[61] Mr Miller also did not find claim 11 an inventive step. Claim 11 relates to an “apparatus according to any one of preceding claims, wherein the container is closed at the top to reduce oxidation of the solder balls and open at the bottom for direct entry of the balls to the positioning member substantially throughout the bottom of the container”. According to Mr Miller, Kasai defines a container, having a slit at the bottom while Ooroku illustrated a container with an open bottom. According to him, it would have been obvious to isolate the solder balls from direct contact with the surrounding atmosphere to reduce the problem of oxidation and contamination of the solder balls.

[62] In relation to claim 12, which concerns an “apparatus according to any one of the claims 6 to 11, wherein the plurality of apertures are located substantially toward one end of the positioning member whereby the container is in communication with the apertures on the positioning member only while the container is in the vicinity of the second end position”, Mr Miller was emphatic that there is nothing inventive about locating the plurality of apertures substantially towards one end of the positioning member whereby the container is in communication with apertures on the positioning member while the container is in the vicinity of the second end position. He added that if the apertures are located at one end of the positioning member, for example the second end position, it is obvious that the container must be in communication with the apertures while being in the vicinity of the second end position for the solder balls to be placed into the apertures.

[63] I agree with Aurigin’s submission that the evidence at the trial showed that ASM’s patent involves nothing more than an aggregation of the known features of a closed top container (as taught by Stumpe), a positioning member located at the end of a template (as taught by Tondelli), a tilting mechanism (as taught by Oguro, Ooroku and Kasai) and a bi-directional tilting mechanism (as taught by Oguro and Kasai). It follows that there was nothing inventive in ASM’s patent.

[64] As I accept the evidence of Aurigin’s expert witnesses, Mr Miller and A/P Tay, that there was nothing novel or inventive in ASM’s patent, I order that this patent be revoked. That being the case, I need not consider whether or not Aurigin’s AU800 infringed ASM’s patent.

[65] As for Aurigin’s counterclaim with respect to groundless threats, reference should first be made to s 77 of the Act, which provides as follows:

[66] ASM had demanded through its solicitors on 6 December 2006 that Aurigin cease and desist from “making, disposing of, offering to dispose of, importing, or keeping whether for disposal or otherwise, as well as withdraw from circulation and sale of the infringing product” in order to avoid legal proceedings. This constitutes a threat of legal proceedings within the meaning of s 77(1) of the Act and Aurigin, being the aggrieved party, is entitled to relief unless ASM is able to justify the threat. Such a threat may be justified under s 77(2) if ASM is able to satisfy two conditions. First, it must be shown that the acts in respect of which ASM had threatened proceedings constitute or would constitute an infringement of its patent. Secondly, the patent must not be shown to be invalid in a relevant respect. The use of the word “and” in s 77(2) shows that both conditions must be satisfied.

[67] As regards the second condition, where the patent of the party making the threats is found to be invalid, its threats are unjustified and the aggrieved party is entitled to relief. In Siegfried Demel (Trading As Demotec Siegfried Demel) v. C & H Jefferson (A Firm) and Another [1999] FSR 204 (“Demel v Jefferson”), a case concerning s 70 of the United Kingdom (“UK”) Patents Act 1977 (c 37), on which s 77 of the Act is based, the defendant had alleged that the plaintiff had violated two of the defendant’s patents (for inventions concerning prostheses for cows) and threatened legal proceedings. The plaintiff brought a claim for groundless threats of legal proceedings and, in that process, challenged the validity of the defendant’s patents and sought an order for revocation of the patents. The defendant resisted the revocation of one of its patents but not the other. The court held that the consequence of this was that the threats made in respect of the patent which revocation had not been challenged were not justified. The plaintiff was thus entitled to relief in respect of the threats pertaining to the invalid patent.

[68] It may be noted that the UK position has changed after 2005 because of the inclusion of s 70(2A) of the UK Patents Act, which provides as follows:

[69] As a result of the inclusion of s 70(2A), the present position in the UK is that even if the patent of the party making the threats is shown to be invalid, that party may avoid liability for making the threats by showing that when the threats were made, he did not know, and had no reason to suspect, that the patent was invalid. There has, however, been no such modification to s 77 of the Act. As a result, the Singapore position is that explained above (at [66]) and illustrated in Demel v Jefferson, which precedes the 2005 amendments to the UK Patents Act.